I. Clinical and Genomic Studies of Uveal Coloboma Since initiating this research, I have recruited and examined approximately 146 famlies (562 participants) where at least one member is affected by uveal coloboma . All probands and their first degree relatives (when available) have complete ophthalmic exams. General physical examinations and targeted systemic testing (e.g., renal ultrasounds, echocardiograms) were performed on probands, as needed. Lymphoblastoid cell lines were established on all participants for candidate gene analysis. As noted in last year's report, we have established a custom-capture, high-throughput sequencing platform for evaluating our coloboma patients for mutations in known coloboma genes, genes that appeared from our unbiased screen using laser captured microdissected tissue in mouse, and genes known to cause coloboma in animal models. To date, we have analyzed 72 trios. As with our pilot experiments, we have demonstrated that the yield on molecular testing for this largely non-syndromic coloboma cohort for known genes is <5%. (This includes back-filling of exons not well-captured by our design with PCR/Sanger sequencing.) However, our screens of candidate genes have led us to pursue experiments on several interesting leads. As part of our whole exome/whole genome protocol, we were awarded a competitive grant for sequencing of 75 exomes through the NIH Clinical Center Genomics Opportunity. We have enrolled a total of 58 participants with largely syndromic coloboma and other developmental eye defects and are currently evaluating data as they become available. II. Laboratory Studies of Uveal Coloboma A. Mouse Models of Coloboma. 1. A novel Pax2 mutant mouse model of coloboma. My lab identified and characterized a mouse model of autosomal dominant congenital optic nerve excavation caused by a missense mutation predicted to change a highly-conserved threonine to alanine in the paired domain of the Pax2 gene. Pax2 is dynamically expressed at the closing edges of the optic fissure and homozygous mutation results in uveal coloboma. Details of our characterization of this model and the Pax2 mutation have been published in PLoS Genetics. Since last annual report, my lab re-attempted to recapitulate the developmental profiling experiments we performed in wild-type mice (Brown, PNAS) in Pax2 mutant embryos (wild-type vs. heterozygous vs. homozygous mutant) across the three developmental time points for optic fissure closure (E10.5-E12.5). All samples have been collected and sent for sequencing through the NIH Intramural Sequencing Center (NISC) and will be analyzed using our in-house pipeline. 2. The RICO Mouse Model of Coloboma The RICO mouse arose from the random insertion of a transgene (NSE-VEGF) on chromosome 13 of C57BL/6 mice. Since the time of the last report, we have identified the junctional fragments of the insertion using whole genome sequencing. We have identified that transgene insertion (30 copies) was associated with an inversion, three duplications and a deletion in a gene desert. We have confirmed the inversion with FISH. We have worked out an improved method of genotyping embryos and have evaluated the effect of the RICO mutant on the expression of genes in the region. We have isolated eye tissue from RICO and control embryos at E11.5 and are exploring differences in chromatin availability using ATAC-Seq. B. Identification of coloboma candidate genes by molecular characterization of gene expression during optic fissure closure. 1. Zfp503 and Zfp703 Our previous work, published in PNAS, identiified two zinc-finger motif-containing genes, Zfp703 and Zfp503 to be important in regulating optic fissure closure in zebrafish. We have created knockout mice for both Zfp703 and Zfp503 and documented germline transmission, homologous recombination andd documented embryonic lethality for homozygous mutants in both cases. The Nlz1 mice do not breed well and the embryos we have collected show a syndromic phenotype that we are trying to characterize. Morphologically, the Nlz2 knockout embryos are considerably more normal and have been observed as late as E18.5. Both models exhibit coloboma. Since the last annual report, we have made greater progress on the Nlz2 mouse, due to the aforementioned breeding problems with the Nlz1 mouse. In addition to characterizing the mechanism of coloboma formation with several relevant candidate markers known to be important in eye development, we have conducted RNA-Seq on RPE samples from wild-type, heterozygous, and homozygous KO embryos at E11.5. We have analyzed the data and are currently verifying several candidates with roles in cell signaling and extracellular matrix biology. We have screened for mutations in Nlz2 in our coloboma patient DNA samples. In addition, we have made a more careful study of the zfp703 zebrafish morphant phenotype. We have identified that morphant fish have several important phenotypes such as cystic kidneys and abnormalities in heart development. As such, this model likely represents a syndromic form of coloboma. We have screened for mutations in patients and have published our zebrafish findings in Developmental Biology 2. FAT protocadherins Another gene family that was suggested by our laser-capture screen was the FAT protocadherins. As previously described, we have found that Fat1 and Fat4 are the members of this family that are most highly-expressed during embryonic eye development and that homozygous knockout of Fat1--but not Fat4--results in coloboma. We have shown that the coloboma in Fat1-/- mice is not the result of a global patterning defect and that the eyes of these embryos are approximately normal size until the time of optic fissure closure. The rate of cell division in the developing optic cup is mildly elevated compared to wild-type and there is no obvious change in the rate of cell death. Real-time PCR of embryos on a panel of genes has revealed RPE-specific changes in several important cell adhesion molecules and signaling molecules. We have extended these studies using a zebrafish model (which also shows coloboma) to determine the precise molecular mechanism behind FAT1-mediated coloboma. 3. CRISPR-mediated knockout of genes associated with optic fissure closure We are systematically knocking out all of the genes for which a zebrafish phenotype is unknown found on our initial screen in mouse (Brown et al, PNAS) using CRISP technology. We have created a pipeline for candidate validation and for phenotype screening, and have begun our first 15 genes.